def test_get_orf_base(self):
expected = pd.DataFrame(index=range(1))
expected['id'] = 'Single_FA'
expected['aa_sequence'] = 'MIMIKL*P'
expected['frame'] = 1
expected['strand'] = '+'
expected['seq_length'] = 8
expected['seq_length_nt'] = 26
expected['orf_sequence'] = 'MIMIKL'
expected['start_site'] = 1
expected['stop_site'] = 7
expected['orf_length'] = 6
expected['start_site_nt'] = 1
expected['stop_site_nt'] = 21
expected['utr3_length'] = 5
expected['first_MET'] = 'M'
expected['final_stop'] = 'STOP'
expected['isoform_number'] = 1
expected['orf_class'] = 'complete'
expected['fasta_id'] = '>Single_FA.orf1 complete:1-21 strand:+'
all_sequences = read_fasta('test_data/test_getorfs.fa')
orf_df = get_orfs(all_sequences, min_orf_length=5)
self.assertTrue(orf_df.equals(expected))
def test_get_orf_all(self):
expected = pd.DataFrame(index=range(2))
expected['id'] = ['Single_FA', 'Single_FA']
expected['aa_sequence'] = ['MIMIKL*P', 'GLQLNHDH']
expected['frame'] = [1, 3]
expected['strand'] = ['+', '-']
expected['seq_length'] = [8, 8]
expected['seq_length_nt'] = [26, 26]
expected['orf_sequence'] = ['MIMIKL', 'GLQLNHDH']
expected['start_site'] = [1, 1]
expected['stop_site'] = [7, 8]
expected['orf_length'] = [6, 8]
expected['start_site_nt'] = [1, 3]
expected['stop_site_nt'] = [21, 26]
expected['utr3_length'] = [5, 0]
expected['first_MET'] = ['M', 'ALT']
expected['final_stop'] = ['STOP', 'ALT']
expected['isoform_number'] = [1, 2]
expected['orf_class'] = ['complete', 'incomplete']
expected['fasta_id'] = [
'>Single_FA.orf1 complete:1-21 strand:+',
'>Single_FA.orf2 incomplete:3-26 strand:-'
]
all_sequences = read_fasta('test_data/test_getorfs.fa')
orf_df = get_orfs(all_sequences,
min_orf_length=5,
both_strands=True,
all_orfs=True)
self.assertTrue(orf_df.equals(expected))
def test_convert_nt_output_format(self):
# tests that a length 3 tupple output, and each is the correct numpy
# array type
sequences = read_fasta('test_data/test_frames.fa')
ids, aa_frames, frame, strand, seq_length_nt, seq_length = translate_all_frames(
sequences, both_strands=False)
orf_sequence, start_sites, stop_sites, orf_length, last_aa_is_stop = find_longest_orfs(
aa_frames)
# filter data by minimum orf length
keep = orf_length >= 6
aa_frames, frame, strand, seq_length_nt, ids, seq_length, start_sites, stop_sites, orf_sequence, last_aa_is_stop, orf_length = filter_objects(
keep, aa_frames, frame, strand, seq_length_nt, ids, seq_length,
start_sites, stop_sites, orf_sequence, last_aa_is_stop, orf_length)
output = convert_start_stop_to_nt(start_sites, stop_sites,
seq_length_nt, orf_length, frame,
last_aa_is_stop)
t_len = len(output) == 3
# test numpy types of all outputs
t0 = output[0].dtype == 'int64'
t1 = output[1].dtype == 'int64'
t2 = output[2].dtype == 'int64'
all_right_types = t0 and t1 and t2 and t_len
self.assertTrue(all_right_types)
def test_read_fasta(self):
# check that files are read into correct format"
read_sequence = read_fasta('test_data/test_mutliple_frame_orfs.fa')
seq_array = [str(x.seq) for x in read_sequence]
# check sequence matches
# (only check first/last few nts, and total length)
t_start = seq_array[0][0:20] == 'GCTTCGGGTTGGTGTCATGG'
t_end = seq_array[0][-1:-20:-1] == 'AGTTGTGTTACCGGGACGG'
t_len = len(seq_array[0]) == 2757
self.assertTrue(t_start and t_end and t_len)
def test_translate_bothstrands(self):
sequences = read_fasta('test_data/test_trans_all_frames.fa')
ids, aa_frames, frame, strand, seq_length_nt, seq_length = translate_all_frames(
sequences, both_strands=True)
frame_correct = np.all(frame == np.array([1, 1, 2, 2, 3, 3]))
strand_correct = np.all(
strand == np.array(['+', '-', '+', '-', '+', '-']))
trans_correct = np.all(aa_frames == np.array([
'MANATEE*', 'LFFGRVRH', 'WRTRPKN', 'YSSVAFA', 'GERDRRI', 'ILRSRSP'
]))
self.assertTrue(frame_correct and strand_correct and trans_correct)
def test_convert_utr_nt(self):
sequences = read_fasta('test_data/test_frames.fa')
ids, aa_frames, frame, strand, seq_length_nt, seq_length = translate_all_frames(
sequences, both_strands=False)
orf_sequence, start_sites, stop_sites, orf_length, last_aa_is_stop = find_longest_orfs(
aa_frames)
# filter data by minimum orf length
keep = orf_length >= 6
aa_frames, frame, strand, seq_length_nt, ids, seq_length, start_sites, stop_sites, orf_sequence, last_aa_is_stop, orf_length = filter_objects(
keep, aa_frames, frame, strand, seq_length_nt, ids, seq_length,
start_sites, stop_sites, orf_sequence, last_aa_is_stop, orf_length)
start_site_nt, stop_site_nt, utr3_length = convert_start_stop_to_nt(
start_sites, stop_sites, seq_length_nt, orf_length, frame,
last_aa_is_stop)
self.assertTrue(np.all(utr3_length == np.array([5, 4, 3])))
def test_translate_output_format(self):
# tests that a length 3 tupple output, and each is the correct numpy
# array type
sequences = read_fasta('test_data/test_trans_all_frames.fa')
output = translate_all_frames(sequences, both_strands=False)
t_len = len(output) == 6
# test numpy types of all outputs
t0 = output[0].dtype.type == np.str_
t1 = output[1].dtype.type == np.str_
t2 = output[2].dtype == 'int64'
t3 = output[3].dtype.type == np.str_
t4 = output[4].dtype == 'int64'
t5 = output[5].dtype == 'int64'
all_right_types = t0 and t1 and t2 and t3 and t4 and t5 and t_len
self.assertTrue(all_right_types)
def test_translate_seq_length(self):
sequences = read_fasta('test_data/test_trans_all_frames.fa')
ids, aa_frames, frame, strand, seq_length_nt, seq_length = translate_all_frames(
sequences, both_strands=False)
self.assertTrue(np.all(seq_length == np.array([8, 7, 7])))
def test_translate_alltransframes(self):
sequences = read_fasta('test_data/test_trans_all_frames.fa')
ids, aa_frames, frame, strand, seq_length_nt, seq_length = translate_all_frames(
sequences, both_strands=False)
self.assertTrue(
np.all(aa_frames == np.array(['MANATEE*', 'WRTRPKN', 'GERDRRI'])))
